Fwd: st: Ranking derived from sports scores

[Jeff Davies <j@jduk.com>]

Thanks for the guidance, Austin and Phil.

Are you aware of software that can take a database of scores (I have a tennis database with about about 750,000 records) composed of player 1, player2, date, games for, games against (+ some other miscellaneous data) and produce a ranking? If so, what type of anaylsis would be appropriate?

If there is limited choice, are you aware of a suitable algorithm that I can use to create such software?

Many thanks to anyone who can assist.

I will be happy to share any new process with the list.

Jeff


On Jan 5, 2011, at 1:53 PM, Phil Schumm wrote:

> On Jan 5, 2011, at 12:20 PM, Austin Nichols wrote:
>> I think a better Wikipedia entry point is
>> http://en.wikipedia.org/wiki/Elo_rating_system
>> but I maintain that the Glickman refs and discussion I already linked to are a better starting point.
> 
> 
> Indeed -- I simply took a quick look for a publicly accessible description of the Bradley-Terry model (i.e., Googled "Bradley-Terry"), and that's the first link that popped up.  I just skimmed Glickman's "A Comprehensive Guide to Chess Ratings," and that includes a very nice description of the Bradley-Terry model and how it relates to the Elo rating system.
> 
> Thanks for the correction.
> 
> 
> -- Phil
> 
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/



*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/